The vast majority of meteorologists agree that weather forecasts beyond 10 days are unreliable. So when AccuWeather claimed on Friday, a month ahead of a total solar eclipse happening on April 8, to have a “scoop” on the cloud forecast, it didn’t take long for meteorologists to roundly blast the forecast as “irresponsible,” “shameful” and “misleading.”

The millions of people hoping for a clear view of the eclipse are no doubt keenly watching the forecast, especially for clouds, which can make or break the eclipse-viewing experience.

Historical information indicates that, on average, cloud cover is least likely on April 8 along the southwestern portion of what’s known as the path of totality — the narrow stripe from Texas to Maine where people will see the moon completely covering the sun — while cloud cover is most likely along the northeastern portion.

But actual cloud cover is one of the more challenging aspects of weather to predict even just a few days ahead of time, let alone a month out.

“Meteorology has come a long way and long-range predictions are generally more accurate than ever. However, predicting cloud cover at a lead time of 30 days is still out of our reach,” John Homenuk, a meteorologist for New York Metro Weather, said in an email.

Yet AccuWeather said its “long-range forecasters have identified weather patterns and emerging trends to create the first look at the weather forecast for the astronomy event of the decade,” and went on to identify areas that it says have the lowest and highest chance of clouds. The forecast cited “multiple weather factors” including the “possibility that a cold front will be moving across central and eastern U.S. near this time of the eclipse,” “the risk of a slow-moving storm system over the Southeast during April” and the potential impact of a weakening El Niño.

In response to AccuWeather promoting its forecast on X as a “scoop,” meteorologists criticized the longtime private weather company for suggesting a credible 30-day cloud forecast was possible:

• “This is irresponsible meteorology, regardless of how the article couches the approach. There is no ‘scoop.’ This far out, climatology is still the best forecast,” said Jon Nese, a meteorologist and professor at Penn State.
• “A ‘weather forecasting’ company trying to be ‘first’ rather than being ‘accurate’ regarding the eclipse forecast. An irresponsible article that equates into sensationalism media, which has no place within the science of meteorology,” said Colby Pope, a meteorologist at the National Weather Service in Little Rock
• “Downright shameful. You simply can not predict cloud cover a month in advance. The science does not exist,” said Mark Paquette, a meteorologist at Earth Networks.
• “This is very misleading. Unless I’ve missed some very impressive recent developments, cloud cover forecasts at 1-month lead time have essentially no skill,” said Justin Schoof, a meteorologist and professor at Southern Illinois University.
• “This is a horoscope. Not SCIENCE. This company continues to have great difficulty comprehending the difference,” said Dan Satterfield, a meteorologist for WBOC in Salisbury, Md.

The AccuWeather story did express the uncertainty in the forecast, noting that even in areas with the highest risk of clouds, “there is still a chance that cloud-free weather will win out,” and highlighted the overall difficulty of the forecast, explaining that “the solar eclipse cloud forecast is complicated as meteorologists are not predicting if it will be cloudy on April 8, but rather, if it will be cloudy for the four-minute window on April 8 when the moon completely blocks out the sun.”

“The article is loaded with plenty of caveats, so in that case, it’s fine in and of itself. However, as is often the case, the framing of the article via headline or graphic seems to convey a bit more confidence than perhaps exists,” Matt Lanza, managing editor of Space City Weather in Houston, said in an email. “But as the article implies, there are so many factors that contribute to cloud cover, specifically cloud cover that could interfere with eclipse viewing, I don’t think this does much for anyone planning their eclipse day.”

Asked about criticism of the forecast, AccuWeather told The Washington Post that, upon further review, it plans to make changes to its story.

“AccuWeather’s eclipse forecast is based on the long-term historical average cloud cover data … as a foundation, and then our long-range forecasters further identified weather patterns and trends that may potentially affect cloud cover,” Jon Porter, AccuWeather’s chief meteorologist, said in an email to The Post. “Our intent was to provide an early forecast for potential cloud cover on April 8, not strictly for the few minutes of totality. We have taken a second look at our article and we can see how that might have been misinterpreted, and we will be updating our story and graphics [on Monday] to better reflect that.”

This isn’t the first time AccuWeather has received blowback for pushing the forecast envelope. The current criticism is reminiscent of when the State College, Pa.-based company began issuing 45-day weather forecasts in 2013, and then extended its forecast to 90 days in 2016, both well beyond the commonly accepted limits of specific weather forecasts for clouds, temperature, wind and precipitation.

Forecasts “should not imply that extended-range forecasts (i.e., 8 days and beyond) are as reliable as short-range forecasts by offering the same amount of detail. Presently, forecasts of specific weather elements in this range are rarely skillful, meaning they are usually no more accurate than predicting that the current conditions will persist or that average conditions will occur,” according to the American Meteorological Society. “However, forecasts issued to highlight the probability of above average or below average conditions occurring in general can be skillful in this range.”

For those who say that climate forecasts for years or decades into the future can’t be credible when weather forecasts for the next few days aren’t always accurate, scientists say that’s not true, because weather models and climate models are fundamentally different in what they are trying to predict.

“In many ways, it is actually harder to forecast weather 2 weeks in advance than it is to forecast climate decades in advance because weather is about the exact conditions and climate is about average conditions,” according to the U.S. government’s climate.gov website. “And while the exact weather at a given location can change dramatically from hour to hour, the average climate changes much less from year to year or even decade to decade. The difference in time scale means that our ability to predict future climate doesn’t depend on our ability to predict next week’s weather.”

So, when will the eclipse cloud forecast start coming into focus?

“As we go through the next 7-14 days, I think we’ll start to get a better picture of how things will evolve,” Homenuk said. “The actual cloud cover forecast — in other words, where exactly you should go to see the eclipse — likely won’t be pinned down with confidence until 48 hours or less before the event begins.”

The Washington Post is a customer of AccuWeather for weather services and forecasts in its print edition, for predictions no more than 10 days into the future.

Jason Samenow contributed to this report.